| NO_x emissions from coal-fired power plants are a major source for air pollution.In recent years,coal-fired power plants with larger capacity and higher parameters have been widely constructed,and hence the environment proctection agency has imposed more restirctive regulations on pollutant emissions from coal-fired power plants.Selective catalytic reduction(SCR)technology has become a widely-adopted denitrification control method in coal-fired power plants due to its high denitrification efficiency and ignorable impacts on boiler operation.With the rapid development of renewable energy and its large-scale penetration to the grid,coal-fired power plants usually need to perform deep peak regulation to balance the fluctuations of renewable energy penetration.Frequent variaion in unit operation fluctuates the flue gas parameters in the boiler,which further affects the control performance of the SCR system.SCR denitrification system is a large-time delay,strong coupling and nonlinear control object,in which the traditional PID control shows a poor control performance.To meet the emission regulation of NO_x in power plants,this work aims to achieve the optimal control of denitrification system based on the analysis of traditional denitrification control strategies.Based on the mechanism analysis of denitrification reactions and the historical operating data of the power plant Supervisory Information System(SIS),the mechanism model and data model of the SCR system were established.Through the simulation of the traditional PID denitrification control system,the parameters self-tuning fuzzy PID control and the predictive control were introduced to optimize the optimal control of denitrification system,avoid excessive ammonia injection,improve effectiveness and economy of SCR control system.The main research work is as follows:(1)The problems of denitrification control in 1000 MW coal-fired unit were analyzed through the actual operation data distribution.Based on the principle and mechanism of the SCR denitrification,a set of differential equations describing the denitrification process were derived from the perspective of kinetics.The fourth order Runge-Kutta method was used to solve the differential equations,and the unknown parameters were identified by the actual operating data and the particle swarm optimization(PSO)algorithm,then the mechanism model of the SCR system was established,and the model was verified by actual denitrification operating data of a 1000 MW unit.(2)To improve the modeling quality of the data samples and reduce the modeling complexity,wavelet threshold was applied to denoise the modeling data and the principal component analysis was used to reduce the dimension of the input variables.Based on the input and output data set after preprocessing,the data based model of SCR system was established by using adaptive fuzzy neural network system.The validity of the model was verified by the actual operation data and the prediction effect was compared with that of BP neural network model.(3)Through the research on the control strategy of the traditional SCR system,the traditional PID control system was modeled and investigated on the Matlab Simulink platform,and the control effect of the traditional control was analyzed by step response and external disturbance simulation.(4)To overcome the limitations in traditional PID control,a fuzzy PID controller was designed to adjust the parameters of the PID controller.Aiming at the nonlinear and large-time delay characteristics of the SCR denitrification system,the predictive control was introduced to optimize the optimal control of ammonia injection in denitrification system.The simulation results showed that,compared with the traditional PID control method,the predictive control system has a shorter adjust time and avoids excessive ammonia injection while ensuring the control level of NO_x concentration. |
PDF Full Text Request